Good Morning..
I have four host computers which hold a variety of virtual servers. We are seeing intermittent communication slowness when performing some SQL based operations. (magnitude of 100 to 300 times slower than expected) There are three experimental designs I have put together to test three theories. It has been a long time, and though I think I have good data to analyze, I have forgotten much of my college training in Stats.

The first hypothesis concerns host to host SQL throughput. I have 4 hosts (A.B.C.D) and have set up aa,ab,ac,ad,ba,bb,bc,bd,ca,cb,cc,cd,da,db,dc and dd (test initiating host - database host) tests randomized and running the exact same database schema check program. I have some other IV info collected to check other factors, but I am focusing on host to host interaction. The collected data is the time to complete the program.

1 - What tests should I run against the data to determine if any of the interactions are exhibiting root cause problems ( initiating host, database host, initiating x database)

2 -I am trying to get this info crunched in R, but I have never used R before. Any help here would be wonderful as well. (I have gotten the data into a data frame. I am having syntax trouble getting anything out)

The second test is basically the same, only that there are 5 different software versions present throughout the system and the hypothesis is that versions and version interactions have no bearing on throughput time. (5 versions of software, 50 tests {25 tests replicated.}. I am assuming the same analysis can be run for this data set, but I do have replication here which I do not have in the first.

The third test entails unzipping a file from host A to receiving host B measuring time to complete. Tests were run in conjunction with the second test on the 50 experiments.

3- Any thoughts and references are greatly appreciated.